Scribd
Upload
104 views18 pages

Topics: Multiple Regression Analysis (MRA)

The document discusses multiple regression analysis (MRA) including its design requirements, assumptions, and comparisons to simple regression analysis. MRA allows prediction of a dependent variable from two or more independent variables and is used for purposes like prediction, explanation, and theory building. Examples are provided to illustrate key concepts such as the multiple regression model equation, R-squared, testing the significance of R-squared and regression coefficients, and comparing regression models.

Uploaded by

chompoonoot
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
104 views18 pages

Topics: Multiple Regression Analysis (MRA)

The document discusses multiple regression analysis (MRA) including its design requirements, assumptions, and comparisons to simple regression analysis. MRA allows prediction of a dependent variable from two or more independent variables and is used for purposes like prediction, explanation, and theory building. Examples are provided to illustrate key concepts such as the multiple regression model equation, R-squared, testing the significance of R-squared and regression coefficients, and comparing regression models.

Uploaded by

chompoonoot
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 18

Topics:MultipleRegression

Analysis(MRA)
ReviewSimpleRegressionAnalysis
MultipleRegressionAnalysis

Designrequirements
Multipleregressionmodel
R2
TestingR2andbs
Comparingmodels
Comparingstandardizedregressioncoefficients

MultipleRegressionAnalysis(MRA)
Methodforstudyingtherelationship
betweenadependentvariableandtwoor
moreindependentvariables.
Purposes:
Prediction
Explanation
Theorybuilding

DesignRequirements
Onedependentvariable(criterion)
Twoormoreindependentvariables
(predictorvariables).
Samplesize:>=50(atleast10timesas
manycasesasindependentvariables)

Assumptions
Independence:thescoresofanyparticularsubjectare
independentofthescoresofallothersubjects
Normality:inthepopulation,thescoresonthedependent
variablearenormallydistributedforeachofthepossible
combinationsoftheleveloftheXvariables;eachofthe
variablesisnormallydistributed
Homoscedasticity:inthepopulation,thevariancesofthe
dependentvariableforeachofthepossiblecombinationsofthe
levelsoftheXvariablesareequal.
Linearity:Inthepopulation,therelationbetweenthedependent
variableandtheindependentvariableislinearwhenalltheother
independentvariablesareheldconstant.

Simplevs.MultipleRegression
OnedependentvariableY
predictedfromasetof
independentvariables(X1,
X2.Xk)
Oneregressioncoefficientfor
Oneregressioncoefficient
eachindependentvariable
R2:proportionofvariationin
r2:proportionofvariationin
dependentvariableY
dependentvariableY
predictablebysetof
predictablefromX
independentvariables(Xs)
OnedependentvariableY
predictedfromone
independentvariableX

Example:SelfConceptandAcademic
Achievement(N=103)

Example:TheModel
Y=a+b1X1+b2X2+bkXk
Thebsarecalledpartialregression
coefficients
OurexamplePredictingAA:
Y=36.83+(3.52)XASC+(.44)XGSC

PredictedAAforpersonwithGSCof4and
ASCof6
Y=36.83+(3.52)(6)+(.44)(4)=56.23

MultipleCorrelationCoefficient(R)
andCoefficientofMultiple
Determination(R2)
R=themagnitudeoftherelationship
betweenthedependentvariableandthebest
linearcombinationofthepredictorvariables
R2=theproportionofvariationinY
accountedforbythesetofindependent
variables(Xs).

ExplainingVariation:Howmuch?
Predictablevariationby
thecombinationof
independentvariables

TotalVariationinY

Unpredictable
Variation

ProportionofPredictableand
UnpredictableVariation
(1R2)=Unpredictable
(unexplained)variation
inY

Where:
Y=AA
X1=ASC
X2=GSC

X1

R2=Predictable
(explained)
variationinY

X2

VariousSignificanceTests
TestingR2
TestR2throughanFtest
Testofcompetingmodels(differencebetween
R2)throughanFtestofdifferenceofR2s

Testingb
Testofeachpartialregressioncoefficient(b)by
ttests
Comparisonofpartialregressioncoefficients
witheachotherttestofdifferencebetween
standardizedpartialregressioncoefficients()

Example:TestingR2
WhatproportionofvariationinAAcanbe
predictedfromGSCandASC?
ComputeR2:R2=.16(R=.41):16%ofthe
varianceinAAcanbeaccountedforbythe
compositeofGSCandASC

IsR2statisticallysignificantfrom0?
Ftest:Fobserved=9.52,Fcrit(05/2,100)=3.09
RejectH0:inthepopulationthereisa
significantrelationshipbetweenAAandthe
linearcompositeofGSCandASC

Example:ComparingModels
TestingR2
Comparingmodels
Model1:Y=35.37+(3.38)XASC
Model2:Y=36.83+(3.52)XASC+(.44)XGSC

ComputeR2foreachmodel
Model1:R2=r2=.160
Model2:R2=.161

TestdifferencebetweenR2s
Fobs=.119,Fcrit(.05/1,100)=3.94
ConcludethatGSCdoesnotaddsignificantly
toASCinpredictingAA

TestingSignificanceofbs
H0:=0
tobserved=b
standarderrorofb
withNk1df

Example:ttestofb
tobserved=.440/14.24
tobserved=.03
tcritical(.05,2,100)=1.97
Decision:Cannotrejectthenullhypothesis.
Conclusion:ThepopulationforGSCis
notsignificantlydifferentfrom0

ComparingPartialRegression
Coefficients
Whichisthestrongerpredictor?Comparing
bGSCandbASC
Converttostandardizedpartialregression
coefficients(betaweights,s)
GSC=.038
ASC=.417
Onsamescalesocancompare:ASCisstronger
predictorthanGSC

Betaweights(s)canalsobetestedfor
significancewithttests.

DifferentWaysofBuildingRegression
Models
Simultaneous:allindependentvariables
enteredtogether
Stepwise:independentvariablesentered
accordingtosomeorder
Bysizeorcorrelationwithdependentvariable
Inorderofsignificance

Hierarchical:independentvariablesentered
instages

Practice:
Gradesreflectacademicachievement,butalso
studentsefforts,improvement,participation,etc.
Thushypothesizethatbestpredictorofgrades
mightbeacademicachievementandgeneralself
concept.OnceAAandGSChavebeenusedto
predictgrades,academicselfconcept(ASC)isnot
expectedtoimprovethepredictionofgrades(I.e.
notexpectedtoaccountforanyadditional
variationingrades)

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy